G. HOWARD BAILEY 807 



organisms by the ordinary smear method. Thus the same phenomenon of protection 

 in the normal animal is paralleled by what happens when an effective antiserum is 

 employed to prevent or combat a bacterial infection. The same means which cause 

 clumping and removal of bacteria from the blood and their accumulation in the organs 

 also causes an accumulation in the same organs of large numbers of polymorphonu- 

 clear leukocytes which ingest the clumped organisms en masse and bring about their 

 rapid dissolution. 



During his investigations of the effects of antisera in vivo, Bull observed an im- 

 portant relation between the degree of agglutination and the curative action of the 

 sera. Thus in heavy septicemias in rabbits very large doses of the specific serum may 

 actually be detrimental owing to the formation of large clumps of bacteria in the lungs 

 and other organs which thus impede the circulation or prevent phagocytosis. On the 

 other hand, very small doses of serum may not reach certain portions of the body, as 

 the cerebrospinal fluid, in sufficient concentration to prevent subsequent infection. 

 The curative action of certain antisera would therefore appear to have a close rela- 

 tionship to maximum and minimum limits of dosage and the degree of agglutination 

 produced, at least in the case of animals with heavy septicemias. The observation 

 would probably have less significance in man. 



It has already been stated that the degree of agglutination of bacteria in vivo is 

 inversely parallel to the infectiousness of the bacteria for the host. This was well 

 illustrated by BuU in showing the relatively high virulence of Bacillus aviseplicus for 

 rabbits, but not for dogs. It was found that the fresh sera of these animals have no 

 bactericidal effect on this organism, there being a difference in the power of the sera 

 to agglutinate and opsonize the bacteria. A very small quantity of culture produces 

 a fatal septicemia in rabbits, but a subtoxic dose is without effect in dogs. 



PHYSICO-CHEMICAL BASIS OF AGGLUTINATION in vivO 



Just why bacteria should agglutinate so much more readily and rapidly in the 

 circulating blood than in serum in the test tube is not definitely known. Bull suggested 

 that some constituent of the blood of the host aids the specific serum in producing 

 those changes in the bacterial bodies which precede agglutination. Recently Wright' 

 has shown in experiments with India ink that this substance remains as an unaltered 

 suspension in sodium chloride, is slowly precipitated by Ringer's solution, more read- 

 ily by serum and defibrinated blood, and almost instantaneously by whole blood and 

 plasma. Gelatin afforded protection against such precipitation. The more nearly the 

 fluid approaches what is found in the blood stream the more marked is the instability 

 of the suspension and the greater its tendency to precipitate. The most actively 

 precipitating of these fluids (plasma) is probably much less unstable than blood in the 

 vessels with its capacity for clotting uninterfered with by chemical agents. Varia- 

 tion in the effect of blood of different animals has also been noted. 



Since a suspension of bacteria behaves much like an emulsoid as regards precip- 

 itation, Bordet^ considers the effect of agglutinins is to convert the emulsoid sus- 

 pension into a suspensoid condition which should then behave like ink particles. 



'Wright, H. D.: /. Path, b- Bad., 30, 185. 1927. 

 'Bordet, J.: Traite de I'lmmufiUe, p. 297. Paris, 1920. 



